THE UNIVERSITY OF ILLINOIS LIBRARY 630.7 116 b co A6RICULJURAL U1BW \ UNIVERSITY OF ILLINOIS Agricultural Experiment Station BULLETIN NO. 244 THE FEED COST OF MILK AND FAT PRODUCTION AS RELATED TO YIELDS BY H. A. ROSS, H. F. HALL, AND C. S. RHODE URBANA, ILLINOIS, MAY, 1923 SUMMARY OF BULLETIN No. 244 This bulletin presents a method of comparing the feed cost, exclusive of pasture, of milk and of fat production for cows of different annual productions. The study is based upon data obtained from cew-testing association records of 1,605 Holstein cows three years old or over, which were maintained in regions where milk was produced almost entirely for fluid consumption or for condensing. When these records were divided into groups on the basis of the amount of butter fat produced in one 'year, the average production of fat for the groups ranged from 93 to 559 pounds, with an average of 263 pounds for the 1,605 cows. The consumption of concentrates per pound of fat produced was found to be approximately uniform for all groups. The amounts of succulent and dry roughages consumed per pound of fat decreased with an increase in production. The consumption of digestible nutrients per pound of butter fat pro- duced and the annual production of fat per cow were found to be nega- tively correlated ( 0.4570 .0133). With an increase in production, the nutrient consumption per unit of product decreased at an ever-decreasing rate. A theoretical curve was fitted to the observed data, using the formula Y = ' +5.18 ; in which Y = the pounds of digestible nutrients consumed per pound of fat produced, and X = the annual pro- duction of fat in pounds. This formula was used as a basis for comparing the feed cost of fat production for cows of various annual productions. When current values were applied to the amounts of feed consumed by the various groups of cows, it was found that the cost of feed per pound of digestible nutrients was comparatively uniform, altho slightly higher for the highest-producing cows. The maximum variation for any one year of the fifteen-year period, 1908-1922, was eleven-hundredths of a cent, and for the average of the fifteen-year period it was only six- hundredths of a cent. When the 1,605 records were divided into classes based on the amount of milk produced annually, the average prodxiction for the various groups was found to range from 3,081 pounds to 16,711 pounds, with an average for all the cows of 7,506 pounds. The correlation coefficient 0.4180 .0139 indicates the relation be- tween the annual yield of milk and the nutrient consumption per unit of O/C-| r*r\ product. This relation may be expressed by the formula Y = A. -f" oO.YO + 14.95 ; in which Y = the nutrient consumption per 100 pounds of milk produced, and X = the annual production of milk in hundredweight. When the digestible nutrients in the feed consumed by these cows, exclusive of pasture, was compared with the nutrient requirement as com- puted by the Haeckcr Standard, it was found that for the majority of the cows the observed consumption was approximately 65 percent of the theoretical requirement, indicating that 35 percent of the nutrients may have been obtained from pasture. THE FEED COST OF MILK AND FAT PRODUCTION AS RELATED TO YIELDS BY H. A. ROSS, ASSOCIATE IN DAIRY ECONOMY, H. F. HALL, FIRST ASSISTANT IN DAIRY ECONOMY, AND C. S. RHODE, ASSISTANT PROFESSOR OP DAIRY HUSBANDRY EXTENSION That high-producing cows require less feed for the production of a given quantity of milk or fat tha,n do low-producing cows is gen- erally recognized. Upon this fact rests much of the value of the work of cow -testing associations. These associations afford a practical means of determining -the unprofitable cows which may well be elim- inated from the herd. However, only a small percentage of the dairy- men in the country are members of such associations, and most dairy- men, therefore, must rely upon some other means of determining the profitableness of their various cows. Knowledge of the amounts of milk or of butter fat which are produced by the different cows in a herd is, of course, essential to the intelligent culling out of the poor cows ; but even when the yield is known, the problem of determining the minimum production which will return a profit still remains. Since this point varies with the price of feed and labor and the prices of dairy products, it is inadvis- able to attempt to state specifically the amount of milk or fat which must be produced annually per cow in order to obtain a profit. It is possible, however, to compare with some accuracy the cost of feed for cows of one level of production with the cost of feed for cows of vari- ous other levels of production. This study aims to present a method whereby such comparisons can be made upon other than a ' ' dollar and cents" basis, so that it may be used however much the prices of feeds may fluctuate. As a basis for comparing the amounts of feed consumed, the total digestible nutrients in the feeds are used. Necessarily the accuracy of this method of determining relative costs depends on the relative uniformity for all groups of cows of the cost of feed per pound of digestible nutrients. The extent to which this cost varied and its effect upon the accuracy of the method of comparison is discussed later (page 560). COMPARABLENESS OF RECORDS The data on which this study is based were obtained from the records of fifteen cow-testing associations in Illinois. They were col- lected by the associations during the four-year period 1917-1920. The 553 554 BULLETIN No. 244 [May, record for each cow, which covered a period of one year, included the following data for each month: the amount of milk produced, the percentage of fat in the milk, the kind and quantity of feed consumed, and the number of days the cow was pastured. The breed, age, and date of freshening were also given. The estimated weight of the cows was reported in approximately one-half of the records. Thousands of records of individual cows were available for study, but in order that only those which were strictly comparable might be used, careful selection was made as follows : 1. Records were considered from only those associations which were located in regions where milk was produced for fluid consump- tion or for condensing. By thus eliminating the herds producing for a butter-fat market, a fair degree of uniformity in seasonal production was obtained and differences in the rate of feeding resulting from price stimulation were lessened. 2. From these associations, fifteen were selected because of the known exceptional ability of the testers in charge of the work. Eleven of these fifteen men had attended agricultural colleges and all of them were noted for their conscientious attention to detail in test- ing milk and in weighing and recording the feed consumed. Other associations may have had equally able testers, but by taking only those of known ability it was thought that the accuracy of the records would be assured. 3. All records were discarded which were in any way incomplete. Among these were records of cows which had been removed from test before the completion of the year, records which reported the value of the roughages consumed but which failed to state the amounts in all cases, and those records which failed to give the breed or the age of the cows. 4. Records of Holstein cows only were included in the study in order to limit the range in the size of the animals and in the percent- age of fat in the milk. Approximately 19 percent of these cows were pure-breds. 5. Since a significant proportion of the feed consumed by young cows is utilized for growth, records of cows under three years of age were excluded from this study. This method of selection left 1,605 records which were deemed comparable. Slight differences in certain factors, such as the seasonal production and the percentage of fat in the milk, could not be elim- inated, but such factors have been included in the tables so that their effect upon the feed consumption may be estimated. 1983] FEED COST OP MILK AND FAT PRODUCTION 555 QUANTITY OF FEED CONSUMED AS RELATED TO FAT PRODUCTION In order to ascertain the relation between the amount of fat pro- duced per cow per annum and the quantity of feed consumed, the 1,605 records of individual cows were divided into groups on the basis of the amount of the fat production. The average fat produc- tion per cow for the various groups, as shown by Table 1, ranged from 93 to 559 pounds, with a mean of 263 pounds for the 1,605 cows. Altho a wide production range is shown, about 90 percent of all the cows fall in the groups producing between 150 and 375 pounds of butter fat annually per cow, the number of animals in the groups producing amounts above or below these figures being very small. The average fat content of the milk of these cows was slightly higher in those groups of greater productivity, the percentage rang- ing from 3.4 percent to 3.7 percent for the groups having sufficiently large numbers of cows to give comparable averages. Such a slight difference, however, probably meant very little difference in the amount of feed consumed and may therefore be ignored in comparing the nutrient consumption of the different groups of cows. The variation among the different groups in regard to the pro- portionate amounts of fat produced during the pasture period (May to October) are shown in Table 1 in the form of percentages of the total annual production. If the lowest- and highest-producing groups having very few records are omitted, it will be seen that the propor- tion produced during the pasture period decreases as the annual fat production per cow increases. This means, of course, that the lower- producing groups consumed less feed, other than pasture, than would have been the case had they produced as large a proportion of fat in the winter months as did the cows of higher productivity. The extent to which this seasonable variation in production may have affected the feed consumption is indeterminable. FEED EXCLUSIVE OF PASTURE THE BASIS OF COMPARISON The only information available on the amount of feed obtained from pasture by these cows is the number of days the cows were pas- tured during the year. These figures have been included in Table 1, altho they are but a doubtful indication of the amount of nutrients consumed because of the extreme variability in the kind and quan- tity of feed afforded the different animals by the pastures. It should be noted that the number of days the cows were on pasture is great- est for the lowest-producing groups, and that the number decreases with the groups of higher production up to the group producing from 200 to 225 pounds of fat annually. For the cows producing 556 BULLETIN No. 244 [May, TABLE 1. BUTTER FAT PRODUCTION RECORDS OF 1,605 HOLSTEIN Cows THREE YEARS OLD OR OVER Group No. Annual production of fat per cow Number of cows Average production of fat per cow Average fat con- tent of milk Proportion of fat pro- duced during the pasture period, May to October Average length of time cows were on pasture 1 Ibs. 75100 . . 8 Ibs. 93 percent 3 3 percent 53 days 207 2 100125 17 114 3.3 53 218 3 125150 56 138 3.4 59 196 4 150 175 94 161 3 4 53 190 5 175 200 145 188 3 5 48 181 6 200225 171 213 3.5 45 176 7 225250 228 237 3.5 46 174 8 250 275 217 261 3 5 45 175 9 275300 177 287 3 5 42 173 10 300325 . . 172 312 3 6 43 173 11 12 13 325350 350375 375 400 133 90 39 337 361 386 3.7 3.6 3.6 41 39 40 173 169 168 14 400 425 26 412 3.7 41 176 15 425450 15 435 3.7 39 176 16 450475 9 465 3.7 42 156 17 18 475500 500525 3 2 480 517 4.0 4.0 39 35 162 153 19 525550 1 526 3.3 38 61 20 550575 2 559 3.5 30 153 more than this amount, the time on pasture is fairly uniform. There are two possible explanations of this relationship. On the one hand, if the somewhat questionable assumption is made that the number of days on pasture is an index of the total nutrients consumed during that time, it would seem that the poorer cows were forced to obtain their feed from pastures after the better cows had been stabled for the winter. On the other hand, the longer period of pasturing may have been the cause of the lower production by failing to furnish sufficient feed after the close of the regular pasture season. Since in this study there is no way of definitely determining the amount of nutrients obtained from pastures, only the feed consumed other than pasture is considered in comparing the economy of the production of the different cows. AMOUNTS OF FEED CONSUMED The feed, exclusive of pasture, consumed by the 1,605 cows has been summarized under three heads : (1) concentrates, (2) succulent roughage, and (3) dry roughage. Table 2 shows the average amounts 19tS] FEED COST OF MILK AND FAT PRODUCTION 557 TABLE 2. AVERAGE AMOUNTS OF FEED CONSUMED IN ONE YEAR BY 1,605 HOLSTEIN Cows THREE YEARS OLD OR OVER: EXCLUSIVE OF PASTURE As related to fat production Group No. Num- ber of cows Average production of fat per cow Feed consumed per cow Feed consumed per pound of fat produced Con- cen- trates Succu- lent rough- age Dry rough- age Con- cen- trates Succu- lent rough- age Dry rough- age /6s. /6s. /6s. /6s. /6s. /6s. /6s. 1 8 93 688 4 985 1 534 7.4 53.6 16.5 2 17 114 1 231 5 198 1 482 10.8 45.6 13.0 3 56 138 883 6 058 1 628 6.4 43.9 11.8 4 94 161 1 336 6 408 1 546 8.3 39.8 9.6 5 145 188 1 391 5 640 1 918 7.4 30.0 10.2 6 171 213 1 576 6 326 2 109 7.4 29.7 9.9 7 228 237 1 754 6 541 2 086 7.4 27.6 8.8 8 217 261 1 827 6 629 2 192 7.0 25.4 8.4 9 177 287 2 038 6 630 2 353 7.1 23.1 8.2 10 172 312 2 309 6 302 2 122 7.4 20.2 6.8 11 133 337 2 292 7 043 2 258 6.8 20.9 6.7 12 90 361 2 744 6 823 2 347 7.6 18.9 6.5 13 39 386 2 856 7 411 2 277 7.4 19.2 5.9 14 26 412 2 760 7 004 2 060 6.7 17.0 5.0 15 15 435 2 784 5 873 2 219 6.4* 13.5 5.1 16 9 465 3 534 5 673 2 464 7.6 12.2 5.3 17 3 480 2 304 8 880 4 080 4.8 18.5 8.5 18 2 517 3 877 6 980 1 810 7.5 13.5 3.5 19 1 526 4 997 11 835 3 945 9.5 22.5 7.5 20 2 559 4 472 7 547 1 677 8.0 13.5 3.0 of these classes of feeds consumed annually per cow and per pound of fat for the various production groups. The consumption of concentrates bears a consistent relation to the amount of fat produced by the cows in the different groups. In other words, there evidently was a tendency to feed the cows in each herd a uniform amount of grain for each pound of fat produced, regardless of the individual production of the cows. The average amount of concentrates consumed per pound of fat for all the cows whose records were studied is 7.3 pounds, and the average of each group varies but little from the average of all, except in those groups having only a few cows. Succulent roughage consisted almost entirely of corn silage, altho small amounts of green corn, beets, and wet malt were included. If the groups having very few records are omitted, it will be seen that the succulent roughage was consumed in amounts ranging from approxi- mately 6,000 pounds per cow in the lowest-producing groups, to 7,400 pounds per cow in the highest-producing groups. However, the amount of succulent roughage consumed per pound of fat produced decreases rapidly with the increase in production, ranging from 44 558 BULLETIN No. 244 [May, pounds for the low-producing cows to 19 pounds for the high- producing cows. Many of the dairymen managing the herds from which records were obtained fed each cow in the herd the same amount of silage, while others varied the quantity of silage somewhat but not in proportion to the production of milk or butter fat. The relation between dry roughage consumed and pounds of fat produced is similar to the relation between silage consumed and fat produced. The amount of dry roughage ranges from 1,600 pounds to 2,300 pounds per cow, or from approximately 12 to 6 pounds per pound of fat. DIGESTIBLE NUTRIENTS IN FEED CONSUMED Three facts concerning the feed consumption of these cows are apparent from the data in Table 2. These are: (1) Approximately the same amount of concentrates, per pound of fat produced, was consumed by the high-producing cows as was consumed by the low- producing cows. (2) The amount of succulent roughage consumed, per pound of fat produced, decreased as the production per cow increased. (3) In like manner, the consumption of dry roughage, per pound of fat prqduced, decreased with the increase in production. These facts indicate that, in so far as feed consumption is concerned, high-producing cows are more economical producers of fat than are low-producing cows. However, if a quantitative comparison of the groups is to be made, it is necessary to reduce the three classes of feed to a common unit. In this study, total digestible nutrients have been selected as the basis of comparison. Using the analyses given by Henry and Morrison, 1 the amounts of total digestible nutrients in the feeds consumed by each cow indi- vidually were computed. In the case of the various grains, rough- ages, and common by-products, these values were applied directly, but for certain commercial mixed feeds the computation of the amount of digestible nutrients involved some estimates. The list of grains and by-products used in compounding these feeds was available, but the proportions in which they were combined were not known. A formula for each feed, based upon the list of ingredients and upon the protein, carbohydrate, fat, and fiber content as guaranteed by the manufacturer was therefore assumed, and the amount of digest- ible nutrients in the mixed feed was then computed upon the basis of the assumed formula. Such a method was considered sufficiently accurate for a study of this kind, and it is probable that no more error is involved than is involved in the assumption of uniform quality of the grains and roughages (which is made when average values are used in computing the digestible nutrient content). The amounts of these mixed feeds were comparatively small, and they were scattered 'Feeds and Feeding. 15th ed. (1915), pp. 653-666. 19X3} FEED COST OF MILK AND FAT PRODUCTION 559 thruout the various groups, so that any errors involved would have practically no bearing upon the relative economy of the fat production of the various groups. In Table 3 are shown the amounts of digestible nutrients in the concentrates, in the succulent roughage, in the dry roughage, and in the total feed consumed, for each pound of fat produced. There is a significant negative correlation ( 0.4570 .0133) between the annual production of butter fat per cow and the nutrient consump- tion per pound of fat produced. The total digestible nutrients con- sumed per pound of fat produced range from approximately 22 pounds in the two lowest-producing groups to 10 pounds in the groups with the highest production. The decrease in the consumption of nutrients per pound of fat produced, with the increase in production 2584.22 per cow, is expressed by the formula, Y = 4- 5.18 ; in which Y = the pounds of digestible nutrients consumed per -pound of fat produced, and X= the total production of fat during the TABLE 3. DIGESTIBLE NUTRIENTS IN FEED CONSUMED IN ONE YEAR BY 1,605 HOLSTEIN Cows THREE YEARS OLD OR OVER: EXCLUSIVE OF PASTURE As related to fat production Group No. Num- ber of cows Average production of fat per cow Digestible nutrients consumed per pound of fat produced Concen- trates Succulent roughage Dry roughage All feed Observed Computed 1 Ibs. Ibs. Ibs. Ibs. Ibs. Ibs. 1 8 93 5.3 9.4 8.1 22.8 23.5 2 17 114 7.7 8.1 6.1 21.9 21.1 3 56 138 4.6 7.9 5.8 18.3 19.1 4 94 161 6.0 7.0 4.6 17.6 17.5 5 145 188 5.5 5.3 4.6 15.4 16.1 6 171 213 5.5 52 4.7 15.4 15.1 7 228 237 5.5 4.8 4.0 14.3 14.3 8 217 261 5.3 4.5 3.9 13.7 13.5 9 177 287 5.3 4.1 3.8 13.2 12.9 10 172 312 5.6 3.6 3.1 12.3 12.4 11 133 337 5.1 3.6 3.2 11.9 11.9 12 90 361 5.7 3.4 3.1 12.2 11.5 13 39 386 5.6 3.3 2.8 11.7 11.1 14 26 412 5.1 2.9 2.4 10.4 10.8 15 15 435 4.8 2.4 2.6 9.8 10.5 16 9 465 5.5 2.1 2.5 10.1 10.2 17 3 480 3.8 33 3.5 10.6 10 1 18 2 517 5.5 2.3 1.7 9.5 9.8 19 1 526 7.2 4.1 3.0 14.3 9.7 20 2 559 6.2 2.5 1.5 10 2 9.4 'From fitted curve. .560 BULLETIN No. 244 [May, year in pounds. 1 As shown by Fig. 1, the theoretical curve fits closely the actual data for all groups except Group 19, which includes but one animal. Hence it may be concluded that within the limits of the production range shown here, the amount of nutrients consumed per pound of fat produced decreases at an ever-decreasing rate as the production per cow increases. Stated in other words, the curve indi- cates that as a dairyman increases the potential production ability of his herd he decreases the feed cost of producing a pound of fat ; and furthermore, this decrease is more rapid in going from a very low production to the average, than it is in going from the average to a high production. \ X ual Proc/uctton of Fat r>*r Co*r FIG. 1. THE CONSUMPTION OF DIGESTIBLE NUTRIENTS AS RELATED TO FAT PRODUCTION (See Table 3) The high-producing cows in this study consumed a greater pro- portion of concentrates than did the low-producing cows; and since the cost of concentrates per pound of nutrients is normally higher than the cost of roughages, it might appear that to use the nutrient consumption as a basis of comparing feed costs would involve a serious error. This would be particularly true if the records of cows pro- 1 The formulas used in this study were derived by the use of the straight line method. Running, Theodore R. Empirical Formulas, pp. 53-56, 1917. FEED COST OF MILK AND FAT PRODUCTION 561 dueing for a butter-fat market had been included in this study, be- cause the majority of such cows produce the greater part of their milk on pasture and are fed largely on cheap roughages during the winter. However, as was stated previously (page 554), the records of those cows only which were producing for a whole-milk, or year- round, market were included in the study. Altho these cows were not, of course, all managed in the same way, the variation in the cost of feed per pound of nutrients was less than would have been the case had cows in butter-fat regions been included. In order to determine the variation in the cost of the rations of these cows, the amounts of the various kinds of feed consumed were tabulated and the cost of feed per pound of digestible nutrients was computed for each group for each year of the fifteen-year period 1908-1922. The computations were based upon the current values of the feeds consumed. Corn, oats, barley, and hay were figured at the average of Illinois farm values for each month; silage was valued at its hay and corn equivalent. 1 Bran, cottonseed meal, linseed oil meal, middlings, hominy feed, gluten feed, and commercial mixed feed prices were figured at the average daily Chicago wholesale prices, plus sums representing the cost of freight, the dealer's profit, and the cost of hauling the feed to the farm. 2 Table 4 shows the slight variation in the cost of nutrients for each of the nine largest groups (Nos. 4 to 12). The other groups are omitted because the number of cows included is insufficient to give comparable averages. It may be seen from this table that even dur- ing the years 1917 to 1921, when the cost of concentrates was rela- tively much higher than the cost of roughages, the greatest difference in the cost of feed for any two groups amounted to only about one- tenth of a cent per pound of digestible nutrients. The maximum variation between any two groups in the average cost per pound of nutrients for the entire fifteen years was only six hundredths of a cent. In other words, the average cost of feed consumed by Group 12, per pound of fat produced, is found to be 21.0 cents (11.5 X 1.83). If the cost of this same group were computed on the basis of the cheapest ration (Group 6), the cost per pound of fat would be 20.4 cents (11.5 X 1-77), a difference of less than one cent. It should be pointed out that the cost of the feeds for each year was computed from the proportionate amounts of the various kinds consumed during the period of the study, 1917 to 1920; and that these proportions might have been somewhat different in the earlier 1 Pearson, F. A., and Gaines, W. L. The Evaluation of Corn Silage, (un- published data). "The total amounts per ton added to the wholesale price are as follows: 1908 to 1915, $4.50; 1916, $5.50; 1917 and 1918, $7.00; 1919, $8.00; 1920. $8.50; 1921, $7.00; and 1922, $6.00. 562 BULLETIN No. 244 [May, E P O K O s gfe o O U fcg l|g& -^ * 3 OC (N ^O (^ ^ 0J I-H co 10 co co "HOOOO (M CO CO 1^ t^ 00000 8 S oi 1*= s" - [V] Jrf aj 8?- >E 1|S& o ll l| li >-' Q) ^i s I ^2 OTt CO l^. CO 00 * 1C O O CO COO*O CO CO CO * CO CO CO CO 00 .^3 J 03 a S "O 3 O S i-i I-H CO (M (N 11 CQO *-> bo a S" a? W3 T}H ^ CO 1-1 |j * in O OS t-- CO C lO O O5 T}< 1C iCCCO CO * C 1C CO CO rf CO CO CO 3 O.4J fl Jd- = "11 O i-l i-t CO CN CN i-H ^S 4 O 09 o S > ' I3J Og r-< O --H O 1-H g * >O O O5 t^ CO ^ ^ ^ t^- ^ 1C 1C 1C CO C i-H COCO t-' CO * COCO CO Or^-- i^i :>sj S i-l i-H CO CNcN i-H Q< v% 05 S CO rH i-l 00^02 |J T(< IOO500CO 00 ic ^ CO 1C O5 OO ?D l^fMCOCOCO CO C >C 1C CO * C CO IO> 2 i-H i-H C C CO -* d (MO5O CO-* COIN CO t^ t^. 33^ "S-sl 2 S S ^H i-H C* CM C^ 1-H _o 3 ^'"^ 00 >C QO -H fO Q t>-Tfl ^ -^ lOOOOCO CO ^H T-H CO t^ CO C tCCCO * > S O O OS CO (M CO (M 1C 00 rfi 1C 1C *O CO 1C * Tf T-H (N CO -* CO CO CO 1 1 00 03 a e ^^ ^>-G -t u +s ^ ^ i-H ^HCO +t ^TS t* > ou O) O) O) O) O) Oi Oi Ci O5 O5 Ci Oi G5 G2 G^ D 19tS\ FEED COST OF MILK AND FA.T PRODUCTION 563 years because of the varying prices of concentrates and roughages. It would appear, however, that the nutrient basis of comparison is sufficiently accurate for the practical purposes of this study. RELATIVE FEED COST OF BUTTER FAT PRODUCTION Since, therefore, the cost of nutrients in the rations of dairy cows of varying fat production, appears to be fairly uniform, the feed cost of the fat produced may be compared by the use of the preced- ing formula. In Table 5 are presented several such comparisons. The relative nutrient consumption indicates, in a general way, the relative feed cost, which is expressed here in terms of percentages, the nutrients consumed by a group of cows with an average produc- tion of 350 pounds of fat being used as a base. It may be noted that the feed cost of producing a pound of fat is twice as great with cows of but 100 pounds annual production as it is with cows of 350 pounds annual production. Altho it is evident that the cost of feed for any one cow or for any one herd of cows may vary widely from the average, it would seem that these data show, with considerable accuracy, the relative feed cost of producing fat with cows of various production levels. TABLE 5. RELATIVE FEED COST OP PRODUCING ONE POUND OP FAT WITH Cows OF DIFFERENT ANNUAL PRODUCTIONS Based upon digestible nutrient consumption exclusive of pasture Annual production of fat per cow 'Digestible nutrients consumed per pound of fat produced Relative feed cost per pound of fat produced Ibs. 350 Ibs. 11.67 percent 100 325 12 11 104 300 12 60 108 275 13.18 113 250 13.85 119 225 14.64 125 200 15.60 134 175 16.77 144 150 18.23 156 125 20.11 172 100 22.63 194 'Computed from the fitted curve. QUANTITY OF FEED CONSUMED AS RELATED TO MILK PRODUCTION In order to determine the relative feed cost of the milk produced by these cows, the 1,605 records were divided into groups on the basis of the yearly milk production per cow. The average production of the various groups, as shown by Table 6, ranged from 3,081 pounds 564 BULLETIN No. 244 [May, to 16,711 pounds, with an average of 7,506. When the cows were grouped on this basis, the average percentage of fat in the milk pro- duced by the various groups was found to decrease slightly with the increase in production, ranging from 3.6 percent to 3.3 percent for the groups of sufficient size to be considered. Probably, slightly less feed was required for the production of 100 pounds of milk with the lower fat content than was required for the production of the higher testing milk. However, the difference in the amount of feed consumed that may be attributed to the variation in the fat content of the milk was undoubtedly less than the difference in feed consump- tion due to the seasonal variation in production. The higher- producing cows produced a greater proportion of milk during the winter months than did the lower-producing cows. This means, of course, that they consumed more feed, other than pasture, than they would have consumed had they produced the same proportion of milk during the winter months as did the lower-producing cows. These two factors affecting feed consumption (variation in fat content of milk and differences in seasonal production) cannot be eliminated, but as the two tend to counterbalance, the error introduced in the study is thereby lessened. TABLE 6. MILK PRODUCTION RECORDS OF 1,605 HOLSTEIN Cows THREE YEARS OLD OR OVER Proportion Annual Average Average fat of milk pro- Average Group production Number production content duced during length of No. of milk of cows of milk of the pasture time cows per cow per cow milk period, May were on to October pasture Ibs. Ibs. percent percent days 1 2 500 3 500 21 3 081 3.6 49 205 2 3 500 4 500 79 4 094 3.6 57 200 3 4 500 5 500 183 5 065 3.6 50 186 4 5 500 6 500 254 6 032 3.6 46 176 5 6 500 7 500 309 7 013 3.6 45 175 6 7 500 8 500 266 7 972 3.5 43 171 7 8 500 9 500 221 .8 956 3.5 42 170 8 9 50010 500 137 9 938 3.4 41 175 9 10 50011 500 76 10 921 3.4 41 163 10 11 50012 500 33 11 943 3.3 39 172 11 12 50013 500 19 12 865 3.3 41 155 12 13 50014 500 4 13 710 3.3 38 175 13 14 50015 500 1 15 389 3.4 33 152 14 15 50016 500 1 15 825 3.3 38 63 15 16 50017 500 1 16 711 3.4 28 152 FEED CONSUMED AND TOTAL DIGESTIBLE NUTRIENTS IN FEED Table 7 shows the amount of feed consumed both per cow and per 100 pounds of milk produced. Practically the same relation obtains here as was found when the comparison was made on a butter-fat 19SS} FEED COST OF MILK AND FAT PRODUCTION 565 TABLE 7. AVERAGE AMOUNTS or FEED CONSUMED IN ONE YEAR BY 1,605 HOLSTEIN Cows THREE YEARS OLD OR OVER: EXCLUSIVE OF PASTURE As related to milk production Feed consumed Feed consumed per 100 Num- Average per cow pounds of milk produced Group No. ber of cows production of milk per cow Con- cen- trates Succu- lent rough- age Dry rough- age Con- cen- trates Succu- lent rough- age Dry rough- age /6s. Ibs. Ibs. Ibs. Ibs. Ibs. Ibs. 1 21 3 081 746 5 318 1 500 24.2 172.6 48.7 2 79 4 094 1 060 5 764 1 498 25.9 140.8 36.6 3 183 5 065 1 352 5 926 1 783 26.7 117.0 35.2 4 254 6 032 1 562 6 279 1 997 25.9 104.1 33.1 5 309 7 013 1 767 6 494 2 167 25.2 92.6 30.9 6 266 7 972 1 993 6 744 2 280 25.0 84.6 28.6 7 221 8 956 2 230 6 762 2 311 24.9 75.5 25.8 8 137 9 938 2 564 6 748 2 365 25.8 67.9 23.8 9 76 10 921 2 829 6 684 2 184 25.9 61.2 20.0 10 33 11 943 3 117 6 139 2 054 26.1 51.4 17 2 11 19 12 865 3 769 6 600 2 522 29.3 51.3 19.6 12 4 13 710 2 989 4 456 2 399 21.8 32.5 17.5 13 1 15 389 4 386 8 464 1 924 28.5 55.0 12.5 14 1 15 825 4 985 11 869 3 561 31.5 75.0 22.5 15 1 16 711 4 261 5 849 2 089 25.5 35.0 12.5 TABLE 8. DIGESTIBLE NUTRIENTS IN FEED CONSUMED IN ONE YEAR BY 1,605 HOLSTEIN Cows THREE YEARS OLD OR OVER: EXCLUSIVE OF PASTURE As related to milk production Group No. Num- ber of cows Average production of milk per cow Digestible nutrients consumed per 100 pounds of milk produced Concen- trates Succulent roughage Dry roughage All feed Observed Computed 1 Ibs. Ibs. Ibs. Ibs. fas. Ibs. 1 21 3 081 17.1 30.1 23.7 70.9 69.0 2 79 4 094 18.3 25.5 17.6 61.4 62.0 3 183 5 065 19.2 20.6 16.5 56.3 56.7 4 254 6 032 19.1 18.5 15.5 53.1 52.6 5 309 7 013 18.6 16.6 14.5 49.7 49.1 6 266 7 972 18.8 15.0 13.2 47.0 46.3 7 221 8 956 19.0 13.4 11.9 44.3 43.9 8 137 9 938 19.3 12.0 10.8 42.1 41.8 9 76 10 921 19.6 10.8 9.7 40.1 40.0 10 33 11 943 19.7 9.0 8.3 37.0 38.3 11 19 12 865 22.5 8.8 9.2 40.5 37.0 12 4 13 710 15.8 6.0 7.5 29.3 36.0 13 1 15 389 23.9 10.0 5.3 39.2 34.1 14 1 15 825 22.5 13.5 10.5 46.5 33.7 15 1 16 711 19.5 7.5 4.5 31.5 32.9 'From fitted curve. 566 BULLETIN No. 244 [May, basis. The amount of concentrates consumed per 100 pounds of milk produced is approximately constant for all groups, while the amounts of succulent and dry roughages per unit of product decrease as the production increases. The tetal digestible nutrients in the feed consumed per 100 pounds of milk produced (Table 8) decrease at an ever-decreasing rate with the increase in production, in a manner similar to the decrease in nutrients with the increase in fat production (Table 3 and Fig. 1). The correlation between milk production and nutrient consumption per hundredweight of milk is 0.4180 .0139, as compared with the fat production and nutrient consumption correlation of 0.4570 .0133. The fitted curve shown in Fig. 2 is expressed by the formula Y = ' -4- 14.95 ; in which Y= the pounds of di- .A. -j OO.ID gestible nutrients consumed per 100 pounds of milk produced, and X = the annual production of milk in hundredweight. The relative feed cost of producing milk with cows of different annual productions may be computed by the use of this formula in the same way that the relative feed cost of fat production was com- puted. Table 9 shows such a comparison, the nutrients consumed by cows producing 10,000 pounds annually being used as a base. Any other production would serve equally well as a basis of comparison; the one used is selected merely for illustration. In this case the cost of feed per 100 pounds of milk was 67 percent greater for the 3000- pound cows than for the 10,000 pound cows. I i I I I Q ,0 \ Annual Praducf/on of Milk p*r Co* FIG. 2. THE CONSUMPTION OF DIGESTIBLE NUTRIENTS AS BELATED TO MILK PRODUCTION (See Table 8) FEED COST OF MILK AJCD FAT PRODUCTION 567 TABLE 9. RELATIVE FEED COST OF PRODUCING 100 POUNDS OF MILK WITH Cows OF DIFFERENT ANNUAL PRODUCTIONS Based upon digestible nutrient consumption exclusive of pasture Annual production of milk per cow Digestible nutrients consumed per 100 pounds of milk produced Relative feed cost per 100 pounds of milk produced Ibs. 10 000 Ibs. 41.7 percent 100 9 000 43 8 105 8 000 46.2 111 7 000 49.2 118 6 000 52.7 126 5 000 57.0 137 4 000 62 5 150 3 000 69.7 167 'Computed from the fitted curve. NUTRIENTS OBTAINED FROM PASTURE Altho the data in this study do not include figures on the amount of nutrients obtained from pasture, these figures may be computed by an indirect method. Such a method involves the assumption that the difference between the observed nutrient consumption of these cows and the nutrient requirement as computed by a feeding standard represents the nutrients obtained from pasture. This is, of course, rather a broad assumption; but in view of the lack of information concerning the proportion of feed that is obtained from pasture under practical farm conditions, it seems worth while to present these data for whatever they are worth. The digestible nutrient requirement per 100 pounds of milk was computed according to the Haecker Feeding Standard, 3 account being taken of the weights of these cows and of the fat content of the milk. From this computed requirement were deducted the nutrients in the feed consumed exclusive of pasture, which are given in Table 8. The result is the computed amount of nutrients obtained from pasture. Table 10 and Fig. 3 show that the low-producing cows obtained a much greater proportion of their feed from pasture than did the high-producing cows. This conclusion is supported by the fact that the low-producing cows were on pasture a longer time and produced a relatively greater proportion of the milk during the pasture period (Table 11). These data indicate that a surprisingly high percentage of the total nutrients consumed were obtained from pasture. If such a comparison of the feeding standard requirement and the observed nutrient con- sumption is valid, each of the groups of cows obtained over one-third of the feed from this source, altho, for most of the groups, less than See reference bearing this number, page 573. 568 BULLETIN No. 244 [May; 45 percent of the total year's production came in the pasture period from May to October. TABLE 10. COMPARISON OP THE AMOUNT OF NUTRIENTS REQUIRED TO PRODUCE 100 POUNDS OF MILK AS COMPUTED BY THE HAECKER FEEDING STANDARD, AND THE OBSERVED NUTRIENT CONSUMPTION EXCLUSIVE OF PASTURE Total digestible nutrients in feed consumed per 100 pounds of milk produced Group No. Average produc- tion of 'Weight of Age of Average fat content Required by the Haecker 'Consumed by the 1,605 Difference between feeding standard milk per cow cows cows of milk Feeding Standard cows in this study (pasture excluded) require- ment and observed consumption Ibs. Ibs. yrs. percent Ibs. Ibs. Ibs. 1 3 081 968 6.0 3.6 122.2 69.0 53.2 2 4 094 993 5.0 3.6 101.5 62.0 39.5 3 5 065 1 018 5.1 3.6 89.4 56.7 32.7 4 6 032 1 042 5.7 3.6 81.3 52.6 28.7 5 7 013 1 066 5.9 3.6 75.3 49.1 26.2 6 7 972 1 090 6.1 3.5 70.6 46.3 24.3 7 8 956 1 115 6.2 3.5 67.0 43.9 23.1 8 9.938 1 139 6.5 3.4 63.1 41.8 21.3 9 10 921 1 164 6.6 3.4 60.7 40.0 20.7 10 11 943 1 190 7.0 3.3 58.7 38.3 20.4 11 12 865 1 216 6 4 3.3 57.2 37.0 20.2 12 13 710 1 232 7.0 3.3 55.9 36.0 19.9 13 15 389 1 276 7.5 3.4 2 53.9 34.1 19.8 14 15 825 1 287 4.5 3.3 53.4 33.7 19.7 15 16 711 1 309 7.5 3.4 2 52.6 32.9 19.7 'Approximately half the records give the weights of the cows, and the esti- mated weights presented in this table are based upon the averages of these records. 2 Groups 13 and 15 include but one cow each, and in computing the nutrient requirement by the use of the Haecker Standard, a uniform test of 3.3 percent has been used for these two groups. 3 Computed from the fitted curve. 1983] FEED COST OF MILK AND FAT PRODUCTION 569 FIG. 3. AMOUNT OF NUTRIENTS REQUIRED TO PRODUCE 100 POUNDS OF MILK, AS COMPUTED BY THE HAECKER FEEDING STANDARD AND THE OBSERVED NUTRIENT CONSUMPTION EXCLUSIVE OF PASTURE Tho the difference between the observed nutrient consumption and the theo- retical requirement based on Haecker's Standard may not be an exact measure- ment of the amount of nutrients obtained from pasture, it may be taken as a rough approximation. 570 BULLETIN No. 244 [May, TABLE 11. COMPARISON OF THE PROPORTION OF MILK PRODUCED DURING THH PASTURE PERIOD, THE LENGTH OF TIME THE Cows WERE ON PASTURE, AND THE COMPUTED PROPORTION OF NUTRIENTS OBTAINED FROM PASTURE Group No. Production of milk per cow Proportion of milk produced during the pasture period, May to October Average length of time cows were on pasture 'Computed propor- tion of total nutrients consumed during the year, obtained from pasture Ibs. percent days percent 1 3 081 49 205 43.5 2 4 094 57 200 38.9 3 5 065 50 186 36.6 4 6 032 46 176 35.3 5 7 013 45 175 34.8 6 7 972 43 171 34.4 7 8 956 42 170 34.5 8 9 938 41 175 33.8 9 10 921 41 163 34.1 10 11 943 39 172 34.8 11 12 865 41 155 35.3 12 13 710 38 175 35.6 13 15 389 33 152 36.7 14 15 825 38 63 36.9 15 16 711 28 152 37.5 difference between the nutrient requirement per 100 pounds of milk, as computed by the Haecker Standard, and the observed consumption is assumed to represent the nutrients obtained from pasture. FORCED FEEDING FOE MAXIMUM YIELDS AND FEED COST OF PRODUCTION In order to avoid possible misinterpretation in the application of these data on the relative feed cost of milk and fat production, it should perhaps be stated that the study does not show that the most economical production is obtained by feeding for maximum produc- tion. It does show that cows of inherently high production ability are more economical producers than cows of low production ability. A clear distinction should be made between these two factors (rate of feeding and inherent production ability) which tend to determine yield, because the economy of production is quite different in the two cases. If a cow is given increasing amounts of feed, a point is eventually reached where the increased milk flow is not proportional to the in- crease in feed. In feeding for maximum production, this point of greatest efficiency is passed, and the feed consumed per unit of product is usually greater than it would have been had the cow been fed a somewhat lighter ration. This is well illustrated by Holtsmark 's work based on 846 dairy herds in Norway. 4 Table 12, taken from * See reference bearing this number, page 573. 1.9MJ FEED COST OP MILK AND FAT PRODUCTION 571 his study, shows that the yield of milk for each 100 feed units in- creases up to the third group, which received 2,500 feed units. This represents the point of greatest efficiency ; the groups of cows which were fed more than this amount showed decreasing yields of milk for each 100 feed units consumed. TABLE 12. RELATION BETWEEN THE AMOUNT OF FEED CONSUMED AND THE MILK PRODUCED BY 846 DAIRY HERDS IN NORWAY According to Holtsmark Number of feed units consumed per cow Yield of milk per cow Yield of milk per 100 feed units 1 500.. kilograms 923 kilograms 61.5 2 000 1 424 71.2 2 500. 1 813 72 5 3 000 2 131 71.0 3 500 2 399 68.5 4 000 2 632 65.8 4 500 2 837 63.1 In contrast to feeding for maximum production is the building up of a herd of cows with the idea of increasing their potential produc- tion ability. As the herd is improved and 1 the production per cow increased, the relative amount of feed required for maintenance be- comes smaller and smaller and the feed consumed per unit of product continues to decrease, altho at a very slow rate for high-producing cows. Stated in other words, in so far as the feed cost of milk and fat production is concerned, feeding for maximum production is sub- ject to the economic law of increasing costs, while increasing the potential ability for high production is subject to the law of de- creasing costs. The cows included in this study were cared for by practical farmers producing milk for profit. It may be assumed that, with the exception of a few animals which were on official test, these cows were fed with the idea of producing a large amount of milk as economically as possible. This does not mean that each farmer fed to the point of greatest economic efficiency but, since high- and low-producing cows were found in the same herds, it is probable that the means of the larger groups represent cows fed to approximately the same point as regards efficiency in the utilization of feed. It is quite possible that the lowest-producing cows may have been underfed and the extremely high-producing groups, overfed. In general, however, the variation in the production of the cows studied appears to have been due to the inherent differences in production ability rather than to differences in feeding. The data in Tables 5 and 9 may therefore be taken as rep- resenting approximately the relative economy of fat and milk produc- tion by high- and by low-producing cows. 572 BULLETIN No. 244 [May, CONCLUSIONS From this study of 1,605 Holstein cows, three years old or over, it may be concluded that the annual production of milk and fat per cow and the nutrient consumption per unit of product are negatively correlated. As production is increased by increasing the potential production ability of a herd, the amount of nutrients consumed per unit of product decreases at an ever-decreasing rate. This relation is expressed by hyperbolic curves. As long as the cost of nutrients in the rations of dairy cows is fairly uniform, these curves may be used to compare the relative feed cost, exclusive of pasture, of milk and fat production for cows similar to those studied if they are fed and cared for in a like manner. If it is assumed that the difference between the amount of nutrients required, as given by the Haecker Standard, and the amount consumed by these cows in concentrates, succulent roughage, and dry roughage, represents the nutrients obtained from pasture, it would appear that the majority of these cows obtained from pasture approximately 35 percent of the total nutrients consumed. 1923] FEED COST OF MILK AND FAT PRODUCTION 573 BIBLIOGRAPHY 1. BRONSON, W. H. The cost of milk production in Massachusetts. Mass. Agr. Col. Ext. Serv. Bui. 19. 1918. 2. COOPER, MORTON O., BENNETT, C. M., and CHURCH, L. M. A study in the cost of producing milk on four 'dairy farms, located in Wisconsin, Michigan, Pennsylvania, and North Carolina. U. S. Dept. Agr. Bui. 501. 1917. 3. HAECKER, T. L. Investigations in milk production. Minn. Agr. Exp. Sta. Bui. 140. 1914. 4. HOLTSMARK, G. Om Forholdet Mellem Melkeudbytte og Anvcndt Foder. Arch. Math, og Naturvid., 26, 2, 1-17. 1905. 5. HOPPER, H. A., and EOBERTSON, F. E. The cost of milk production. N. Y. (Cornell) Agr. Exp. Sta. Bui. 357. 1915. 6. HOPPER, H. A., BOWEN, H. M., and BARLOW, F. S. Feed consumed in milk production. N. Y. (Cornell) Agr. Exp. Sta. Bui. 398. 1918. 7. MCDOWELL, J. C. Relation of production to income from dairy cows. U. S. Dept. Agr. Bui. 1069. 1922. 8. . Influence of season of freshening on production and income from dairy cows. U. S. Dept. Agr. Bui. 1071. 1922. 9. MENDUM, S. W. Cost of milk production on forty-eight Wisconsin farms. U. S. Dept. Agr. Bui. 1144. 1923. 10. MUSSER, K. B., WHITE, G. C., MCDONALD, B. A., and JUDKINS, H. F. Studies from the survey on the cost of market milk production. Conn. Agr. Col. Ext. Serv. Bui. 7. 1917. 11. RASMUSSEN, F. Cost of milk production. N. H. Col. and Sta. Ext. Bui. 2. 1913. 12. THOMPSON, A. L. Cost of producing milk on 174 farms in Delaware county, New York. N. Y. (Cornell) Agr. Exp. Sta. Bui. 364. 1915. 13. WING, H. H. Cost of milk production, variation in individual cows. N. Y. (Cornell) Agr. Exp. Sta. Bui. 52. 1893. 14. WOLL, F. W. The feed unit system for determining the economy of pro- duction by dairy cows. Wis. Agr. Exp. Sta. Circ. 37. 1912. 15. WOLL, F. W., and HARRIS, R. T. The Wisconsin dairv cow competition. Wis. Agr. Exp. Sta. Bui. 226. 1912. UNIVERSITY OF ILLINOIS-URBANA